Quote Originally Posted by edcculus View Post
Oh, great idea!


One quick question about both Thomas's and Smieglitz's methods. I have my process dialed in a little better. I know for a 35mm negative being printed at 8X10 with a medium aperture (lets say f8), I'm going to require somewhere in the neighborhood of a 30-40 second exposure depending on the negative, crop etc. Unless I have a negative thats really bad, I can rule out any exposure lower than 20 seconds. When I do smaller formats, I even dial down the aperture so I can keep my exposure in the +20 second range. Short exposures seem really hard to control to me, plus its easier to dodge/burn at longer times.
Exactly. I feel the same way about short exposures and the ideal range to be printing at. I always try to print longer, hopefully around 24 to 36 seconds or so. And after thinking about all three testing methods yesterday, I grabbed a difficult negative and ran a test using each method. I earnestly tried to judge the best strip, and then made a quick print from the chosen time. All three strips chosen were at different times. I then scanned the results which are shown below.

When I tried Thomas' method yesterday, I screwed up a couple of attempts when the time got down to 8/5.6/4. But since I knew I didn't want to print at such short times anyways, I ended up using the sequence 45/32/22/16/11/8 instead. Since I usually start with my method at 8 seconds, to make the first strips identical, I thought that sequence would be better for that reason as well. One test used the OP's additive arithmetic method with a constant interval and times of 5/10/15/20/25/30/35/40/45. For my usual method I used 8+2+2.5+3.5+4+5+7+8 (producing 8,10,12.5,16,20,25,32,40-second exposures). Note that the latter two methods have coincident times (20/25) in the middle of the respective sequences. And that's usually as short as I'd ever want to be printing. It also coincides with your lower limit.

I found progressively covering the paper to be a much easier way whether the exposure change was arithmetic (i.e., 5-seconds) or geometric. No surprise there since I normally use the covering (rather than uncovering) method and found out a long time ago that I prefer it. It also seems less confusing to novice students.

Quote Originally Posted by edcculus View Post
With the count down method, I suppose I could make the portions larger, and just stop before I get all the way down to the lowest exposure. With the count up method, would I just expose the entire sheet at the lowest exposure I want to start with (say 16s), then add from there (+4, +5, +7 etc?)?
That's correct. Indeed, another thing I like about my method is that I can start anywhere on one of the ISO film speed numbers and proceed from there. I could begin at 8 or 16 or 25 or 64, etc., and just keep building that sequence. Every third strip will be 1/3-stop different whereas the uncovering sequence is progressing by 1/2 stops. In that method, every other number represents a full-stop difference in exposure, the interval is coarser, and I found the shortest times difficult to do accurately and screwed up twice.

Surprisingly, I think the constant arithmetic interval method led to the better initial work print in this instance. IMO, it was off slightly but mine was very close and the print from the least familiar method came in third. So, I guess I will back up and keep printing with my regular geometric method since it seems much easier for me in practice. Although it was not as accurate in this particular case, I believe it will generally produce better results and I like the fact that each strip shows an equal visual change.

Here they are (not in any particular order, but marked so I can ID the method):

JenX strip test 72.jpg

OK. Time for me to stop beating this horse.